What would happen if we change one simple element of our landing page? How about we add more text or change our pricing? What will happen? These are hypothetical situations that partners and spouses who run businesses together will argue over endlessly and there is absolutely no reason for it. In internet marketing you can have your cake and eat it too. There are some very simple ways to find out what works and what doesn’t. The question is which method to use to get the most useful results.
Enter the world of A/B testing. It’s a concept that’s been around longer than the internet has. It was used with direct mailers to test different offers and designs. It’s been tried in newspaper, magazine and even on billboards in the past to measure response rates. On the internet it’s more technologically advanced but the goal is still the same. Measure two similar products or offers against each other and see which one is going to make you more money.
The pros and cons of A/B testing are also the same as they have always been. In order for the test to produce a true result you can only change one element. Offering two entirely different home pages to the world may tell you which one converts better but it won’t tell you why. The answer you really want to get with A/B testing is which element of the design change is producing the result. A simple color change, additional content, or a price change is a good factor to test in an A/B set-up but all three at once won’t allow you to track which of those is making the difference.
Another drawback to A/B testing is it doesn’t give you any behavioral data. You can run two home pages with a different price on them and track how many units of product sell for each amount but you still will not know why. What is the reason that visitors will buy a product for $40 but not pay $50 for it? What if the result is the opposite and more units sell at the higher price? Is that enough data to make a decision to increase your prices across the board? If you don’t know why the higher price was more successful, that increase could cost you dearly in the long run.
On the pro side, A/B testing is most useful when it is used with other tests that measure elements like behavior, demographics, and economic climate. What works today may not work tomorrow when everyone’s financial situation is a little better (or worse). Knowing why people do what they do is the most valuable piece of information you as an internet marketer can have. Set up a series of tests and pool your results to get a feel for what your market is doing. Call it A through Z testing. It’s far more effective than just a simple A/B.